Extreme-ultraviolet light generating system, and method for manufacturing electronic device

Abstract

To provide an EUV light generation system that can shorten time until the system is restored from an abnormal state.SOLUTION: An EUV light generation system comprises: a pulse laser beam sensor which measures the pulse energy of a main pulse laser beam; a target detection sensor which generates a passage signal of a droplet target, which is used for generating a trigger signal for irradiating a main pulse laser beam; an EUV light sensor which measures the pulse energy of EUV light; and a processor. The processor includes a neural network which uses log data of the pulse energy of the main pulse laser beam, log data of an irradiation pulse interval and log data of the pulse energy of EUV light as input data, and information by which whether or not the EUV light generation system is in any state of a normal state, a connection failure state of the droplet target, an abnormal state of variation of gaps between the droplet targets, or an abnormal state of a relative position between the irradiation position of the main pulse laser beam and a mist-state target can be specified as output data.SELECTED DRAWING: Figure 9

Description

[Technical Field]

[0001] The present disclosure relates to an extreme ultraviolet light generating system and a method for manufacturing an electronic device. [Background technology]

[0002] In recent years, with the miniaturization of semiconductor processes, the miniaturization of transfer patterns in optical lithography for semiconductor processes has progressed rapidly. In the next generation, fine processing of 10 nm or less will be required. For this reason, there is a demand for the development of semiconductor exposure equipment that combines a device for generating extreme ultraviolet (EUV) light with a wavelength of approximately 13 nm and a reduced projection reflective optical system.

[0003] As an EUV light generation device, development of a Laser Produced Plasma (LPP) type device that uses plasma generated by irradiating a target material with laser light is progressing. [Prior art documents] [Patent documents]

[0004] [Patent Document 1] International Publication No. 2022 / 201394 [Patent Document 2] Overview of U.S. Patent No. 10,395,356

[0005] An extreme ultraviolet light generation system according to one aspect of the present disclosure irradiates a droplet target formed by combining a plurality of droplets with a pre-pulse laser beam to generate a mist-like target, and irradiates the mist-like target with a main pulse laser beam to generate plasma, thereby generating extreme ultraviolet light. The extreme ultraviolet light generation system includes: a pulse laser beam sensor that measures the pulse energy of the main pulse laser beam; a target detection sensor that generates a droplet target passage signal for generating a trigger signal for irradiating the main pulse laser beam; and an EUV light sensor that measures the pulse energy of the extreme ultraviolet light. and a processor, the processor including a neural network that receives as input data log data of pulse energy obtained from the pulse laser light sensor, log data of the irradiation pulse interval of the main pulse laser light, and log data of pulse energy obtained from the EUV light sensor, and outputs as output data information that enables identification of whether the extreme ultraviolet light generation system is in a normal state, a state in which the droplet targets are poorly bonded, a state in which the variation in the interval between the droplet targets is abnormal, or a state in which the relative position of the position irradiated with the main pulse laser light and the mist-like target is abnormal.

[0006] According to another aspect of the present disclosure, there is provided an extreme ultraviolet light generation system for generating extreme ultraviolet light by irradiating a droplet target formed by combining a plurality of droplets with a pre-pulse laser beam to generate a mist-like target, and irradiating the mist-like target with a main pulse laser beam to generate plasma, the system including: a pulse laser beam sensor for measuring pulse energy of the main pulse laser beam; a target detection sensor for generating a droplet target passage signal for generating a trigger signal for irradiating the main pulse laser beam; an EUV light sensor for measuring pulse energy of the extreme ultraviolet light; and a processor, wherein the processor measures the pulse energy of the extreme ultraviolet light obtained from the pulse laser beam sensor. the log data of the pulse interval of the main pulse laser beam, and the log data of the pulse energy obtained from the EUV light sensor as input data, and the output data is information that enables identification of whether the extreme ultraviolet light generation system is in a normal state, a state in which the droplet targets are poorly bonded, a state in which there is an abnormal variation in the interval between the droplet targets, or a state in which there is an abnormality in the relative position between the position irradiated with the main pulse laser beam and the mist-like target; generating extreme ultraviolet light by the extreme ultraviolet light generation system; outputting the extreme ultraviolet light to an exposure apparatus; and exposing a photosensitive substrate in the exposure apparatus to the extreme ultraviolet light in order to manufacture an electronic device.

[0007] According to another aspect of the present disclosure, there is provided an extreme ultraviolet light generation system for generating extreme ultraviolet light by irradiating a droplet target formed by combining a plurality of droplets with a pre-pulse laser beam to generate a mist-like target, and irradiating the mist-like target with a main pulse laser beam to generate plasma, the extreme ultraviolet light generation system including: a pulse laser beam sensor for measuring pulse energy of the main pulse laser beam; a target detection sensor for generating a droplet target passage signal for generating a trigger signal for irradiating the main pulse laser beam; an EUV light sensor for measuring pulse energy of the extreme ultraviolet light; and a processor, wherein the processor measures log data of the pulse energy obtained from the pulse laser beam sensor and The method includes: irradiating a mask with extreme ultraviolet light generated by the extreme ultraviolet light generation system, the extreme ultraviolet light generation system including a neural network that receives log data of the irradiation pulse interval of the main pulse laser light and log data of the pulse energy obtained from the EUV light sensor as input data, and that outputs information that enables identification of whether the extreme ultraviolet light generation system is in a normal state, a state in which the droplet targets are poorly bonded, a state in which there is an abnormal variation in the interval between the droplet targets, or a state in which the relative position between the position irradiated with the main pulse laser light and the mist target is abnormal; and exposing and transferring a pattern formed on the selected mask onto a photosensitive substrate. [Brief explanation of the drawings]

[0008] Some embodiments of the present disclosure will now be described, by way of example only, with reference to the accompanying drawings, in which: [Figure 1] FIG. 1 is a diagram showing an EUV light generation system according to a comparative example. [Figure 2] FIG. 2 is a diagram showing a detailed configuration of the target image capturing unit. [Figure 3] FIG. 3 is a diagram showing an imaging area of ​​the target image capturing unit. [Figure 4] FIG. 4 is a diagram for explaining an example of the arrangement of the EUV light sensor. [Figure 5]FIG. 5 is a diagram showing the arrangement of the EUV light sensors as viewed from the X-axis direction. [Figure 6] FIG. 6 is a flowchart showing an example of the operation of an EUV light generation system according to a comparative example. [Figure 7] FIG. 7 is a flowchart showing the details of the processing applied in step S10 of FIG. [Figure 8] FIG. 8 is a flowchart showing the details of the processing applied in step S60 of FIG. [Figure 9] FIG. 9 is a diagram illustrating an EUV light generation system according to the first embodiment. [Figure 10] FIG. 10 is a flowchart illustrating an example of the operation of the EUV light generation system according to the first embodiment. [Figure 11] FIG. 11 is a graph showing an example of log data of the pulse energy of the main pulse laser beam. [Figure 12] FIG. 12 is a graph showing an example of log data of the irradiation pulse interval of the main pulse light. [Figure 13] FIG. 13 is a graph showing an example of log data of EUV light pulse energy detected by one of the EUV light sensors. [Figure 14] FIG. 14 is a flowchart showing the details of the processing applied in step S25 of FIG. [Figure 15] FIG. 15 is a diagram showing a first display example of the estimation result on the terminal. [Figure 16] FIG. 16 is a diagram showing a second display example of the estimation result on the terminal. [Figure 17] FIG. 17 is a flowchart showing the details of the processing applied in step S55 of FIG. [Figure 18] FIG. 18 is a diagram showing the configuration of a learning model used in the first embodiment. [Figure 19] FIG. 19 is a flowchart showing a method for generating a learning model. [Figure 20] FIG. 20 is a diagram showing a schematic configuration of an exposure apparatus connected to an EUV light generation system. [Figure 21] FIG. 21 is a diagram showing a schematic configuration of an inspection device connected to an EUV light generation system. Embodiment

[0009] -table of contents- 1. Overall description of the EUV light generation system according to the comparative example 1.1 Configuration 1.2 Operation 2. Challenges 3. Embodiment 1 3.1 Configuration 3.2 Operation 3.3 How to generate a learning model 3.4 Variations 3.5 Actions and Effects 4. Exposure equipment and inspection equipment 5.Other Hereinafter, embodiments of the present disclosure will be described in detail with reference to the drawings. The embodiments described below show some examples of the present disclosure and do not limit the content of the present disclosure. Furthermore, not all of the configurations and operations described in each embodiment are necessarily essential as the configurations and operations of the present disclosure. Note that the same components are given the same reference symbols, and redundant explanations will be omitted.

[0010] 1. Overall description of the EUV light generation system according to the comparative example 1.1 Configuration FIG. 1 is a diagram illustrating an EUV light generation system 1 according to a comparative example. The comparative example of the present disclosure refers to a configuration that the applicant recognizes as being known only by the applicant, and is not a publicly known example that the applicant acknowledges. The EUV light generation system 1 is an LPP-type EUV light generation system. The EUV light generation system 1 includes a pre-pulse laser device 3P and a main-pulse laser device 3M, and generates a mist-like target by irradiating a droplet-like target with a pre-pulse laser beam 31P output from the pre-pulse laser device 3P. The droplet-like target is referred to as a droplet target 27. The mist-like target is referred to as a mist-like target 274.

[0011] The EUV light generation system 1 generates plasma 275 by irradiating a mist-like target 274 with a main pulse laser beam 31M output from a main pulse laser generator 3M. The generated plasma 275 emits EUV light 277. The EUV light generation system 1 collects the EUV light 277 and outputs it to an exposure apparatus 9, which is an external device of the EUV light generation system 1.

[0012] The pre-pulse laser device 3P is configured, for example, by a YAG laser device or a laser device using Nd:YVO4. The main pulse laser device 3M is configured, for example, by a CO2 laser device. The main pulse laser device 3M may also be configured, for example, by a YAG laser device or a laser device using Nd:YVO4.

[0013] The EUV light generation system 1 further includes a chamber 2, a laser light transmission optical system 33, laser light sensors 501P, 501M, a laser light focusing optical system 22, an EUV light focusing optical system 23, a connection unit 24, a target supplier 25, a stage 26, a target collector 28, a target detection sensor 41, a target image capturing unit 60, a plurality of EUV light sensors 43, and a processor 8.

[0014] The processor 8 is a processing device that includes a CPU (Central Processing Unit) 8C for executing various control programs and a memory 8M in which the various control programs are stored. The processor 8 is specially configured or programmed to execute various processes included in the present disclosure. The processor 8 comprehensively controls the operation of each component of the EUV light generation system 1 based on various commands from an exposure apparatus 9, which is an external device.

[0015] The chamber 2 is a container in which a mist-like target 274 generated inside is irradiated with the main pulse laser beam 31M, thereby generating plasma 275 from the mist-like target 274 and generating EUV light 277. A wall 211 of the chamber 2 forms an internal space of the chamber 2 and isolates the internal space of the chamber 2 from the outside. The wall 211 is provided with a window 215 for introducing the pre-pulse laser beam 31P and the main pulse laser beam 31M into the chamber 2. The chamber 2 also includes a target supply path 212 for supplying a droplet target 27 into the chamber 2.

[0016] The laser beam transmission optical system 33 is an optical system that introduces the pre-pulse laser beam 31P and the main pulse laser beam 31M output from the pre-pulse laser device 3P and the main pulse laser device 3M into the chamber 2 via the window 215. The laser beam transmission optical system 33 is located on the optical paths of the pre-pulse laser beam 31P and the main pulse laser beam 31M outside the chamber 2, and is disposed between the pre-pulse laser device 3P and the main pulse laser device 3M and the window 215.

[0017] The laser beam transmission optical system 33 includes a high-reflection mirror 331P and a combiner 332 that transmit the pre-pulse laser beam 31P, and a high-reflection mirror 331M and a high-reflection mirror 332M that transmit the main pulse laser beam 31M. Each of these optical elements is mounted on a stage (not shown) that adjusts at least one of the position and orientation of these elements. The operation of these stages is controlled by the CPU 8C. The laser beam transmission optical system 33 further includes a beam splitter 500P and a beam splitter 500M.

[0018] The beam splitter 500P is disposed in the optical path of the pre-pulse laser beam 31P, and reflects a portion of the pre-pulse laser beam 31P and transmits the remainder. The laser beam sensor 501P detects the energy of the portion of the pre-pulse laser beam 31P reflected by the beam splitter 500P.

[0019] The beam splitter 500M is disposed in the optical path of the main pulse laser beam 31M, and reflects a portion of the main pulse laser beam 31M and transmits the remainder. The laser beam sensor 501M detects the energy of the portion of the main pulse laser beam 31M reflected by the beam splitter 500M.

[0020] The laser beam focusing optical system 22 is an optical system that focuses the pre-pulse laser beam 31P and the main pulse laser beam 31M introduced into the chamber 2 through the window 215 onto the plasma generation region R1, and is disposed inside the chamber 2. The laser beam focusing optical system 22 includes a laser beam focusing mirror 221 and a manipulator 224.

[0021] The laser beam focusing mirror 221 is mounted on a manipulator 224. The laser beam focusing mirror 221 is configured using an off-axis parabolic mirror 222 and a plane mirror 223.

[0022] The manipulator 224 is a mechanism that adjusts at least one of the position and the attitude of the laser beam focusing mirror 221. The manipulator 224 is a mechanism that adjusts at least one of the position and the attitude of the laser beam focusing mirror 221 so that the droplet target 27 is irradiated with the pre-pulse laser beam 31P and the mist-like target 274 is irradiated with the main pulse laser beam 31M in the plasma generation region R1.

[0023] The driving of the manipulator 224 is controlled by the CPU 8C. The manipulator 224 may be a mechanism that moves the laser beam focusing mirror 221 in a direction along at least one of the X-axis and the Y-axis. The manipulator 224 may be a mechanism that moves the laser beam focusing mirror 221 in a direction along the Z-axis in addition to the X-axis and the Y-axis. The manipulator 224 may be a stage that is a mechanism that adjusts at least one of the position and the attitude of the laser beam focusing mirror 221.

[0024] With regard to the directions of the X-axis, Y-axis, and Z-axis, the direction in which EUV light 277 is guided from the chamber 2 toward the exposure device 9 is defined as the Z-axis direction. The X-axis and Y-axis are orthogonal to the Z-axis and are orthogonal to each other. The Y-axis direction is defined as the direction of the central axis of a nozzle 252 (described later) of the target supply unit 25 that outputs the target material 257 into the chamber 2. The Y-axis direction is the direction of a target trajectory Q (described later).

[0025] The EUV light collecting optical system 23 is an optical system that collects the EUV light 277 and collects it at an intermediate focus point IF. The EUV light collecting optical system 23 is disposed inside the chamber 2. The EUV light collecting optical system 23 includes an EUV light collecting mirror 231.

[0026] The EUV collector mirror 231 reflects EUV light 277 emitted from the plasma 275 in the plasma generation region R1. The EUV collector mirror 231 focuses the reflected EUV light 277 at an intermediate focus point IF located within the connection part 24. The reflective surface of the EUV collector mirror 231 is formed, for example, by a multilayer reflective film in which molybdenum and silicon are alternately stacked. The reflective surface of the EUV collector mirror 231 is formed, for example, by a part of an ellipsoid of revolution having a first focus and a second focus.

[0027] The EUV collector mirror 231 is disposed so that its first focal point is located in the plasma generation region R1 and its second focal point is located at the intermediate focus IF. A through-hole 232 is formed in the center of the EUV collector mirror 231. The through-hole 232 is a hole that allows the pre-pulse laser beam 31P and the main pulse laser beam 31M reflected by the laser collector mirror 221 to pass through toward the plasma generation region R1.

[0028] The connection part 24 is a connection part between the chamber 2 and the exposure device 9. A wall 241 and an EUV shutter 243 are provided inside the connection part 24. An aperture 242 is formed in the wall 241. The aperture 242 is formed so as to be located at the intermediate focus point IF. The EUV shutter 243 is arranged so as to be able to be moved in and out of the optical path of the EUV light 277 so that the output of the EUV light 277 can be adjusted. The opening and closing of the EUV shutter 243 is controlled by the CPU 8C.

[0029] The target supplier 25 is a device that melts a target material 257, which is a metallic material that forms a droplet target 27 supplied into the chamber 2, and outputs the melted target material 257 in the form of droplets toward the plasma generation region R1. The target supplier 25 is a device that outputs the droplet target 27 using a so-called continuous jet method. The droplet target 27 supplied by the target supplier 25 is made of a metallic material. The metallic material that forms the droplet target 27 is a material containing tin, terbium, gadolinium, or a combination of any two or more of these. A preferred metallic material is tin.

[0030] The target supplier 25 is configured using a tank 251, a nozzle 252, a heater 253, a pressure regulator 254, and a piezoelectric element 255. The operation of the target supplier 25 is controlled by a CPU 8C. The target supplier 25 is mounted on a stage .

[0031] The stage 26 is a mechanism for adjusting the position or attitude of the target supplier 25. The stage 26 is a mechanism for moving the target supplier 25 in at least one axial direction of the X-axis, Y-axis, and Z-axis. The stage 26 is a mechanism for adjusting the position of the target supplier 25 so that the droplet targets 27 output from the target supplier 25 are supplied to a predefined target mist-forming region Rmist. Driving of the stage 26 is controlled by the CPU 8C.

[0032] The target collector 28 is a device that collects the droplet targets 27 that have not been irradiated with the pre-pulse laser beam 31P and the main pulse laser beam 31M, among the droplet targets 27 output into the chamber 2. The target collector 28 is provided on the wall 211 of the chamber 2 on an extension of the target trajectory Q.

[0033] The target detection sensor 41 is a sensor that detects the droplet target 27 passing through a target detection area R2. The target detection area R2 is an area at a predetermined position on the target trajectory Q in the target supply path 212. The target detection sensor 41 includes an illumination unit 410 and a detection unit 420.

[0034] The illumination unit 410 and the detection unit 420 are connected to the wall 211 of the chamber 2 that forms the target supply path 212 via the window 216 and the window 217, respectively. The illumination unit 410 and the detection unit 420 are arranged to face each other across a target detection region R2 on the target trajectory Q. The illumination unit 410 and the detection unit 420 are arranged so that the illumination optical axis of the illumination unit 410 and the detection optical axis of the detection unit 420 are approximately coaxial with each other and pass through the target detection region R2, as shown in FIG. 1 . The illumination optical axis of the illumination unit 410 is the optical path axis of the illumination light output from the illumination unit 410 toward the target detection region R2. The detection optical axis of the detection unit 420 is the optical path axis of the illumination light detected by the detection unit 420, of the illumination light output from the illumination unit 410 toward the target detection region R2.

[0035] The illumination unit 410 outputs illumination light toward the target detection region R2 so as to illuminate the droplet target 27 passing through the target detection region R2. The illumination unit 410 is configured using a light source 411 and an illumination optical system 412. The detection unit 420 is electrically connected to the CPU 8C, detects the light intensity of the illumination light output to illuminate the droplet target 27 passing through the target detection region R2, and transmits a detection signal to the CPU 8C. This detection signal may be referred to as a passage timing signal. The detection unit 420 is configured using an optical sensor 421 and a light receiving optical system 422.

[0036] The target image capturing unit 60 captures an image of the droplet target 27 or the mist-like target 274 passing through the target detection region R2 and heading toward the plasma generation region R1.

[0037] The EUV light sensor 43 is a sensor that measures the energy of EUV light 277 emitted from the plasma 275. Each of the multiple EUV light sensors 43 measures the energy of the EUV light 277 from a different direction and transmits the measurement value to the CPU 8C. The operation of each of the multiple EUV light sensors 43 is controlled by the CPU 8C.

[0038] Fig. 2 is a diagram showing a detailed configuration of the target image capturing section 60. Fig. 2 shows the configuration around the target image capturing section 60 when viewed from the -Z direction.

[0039] The target image capturing unit 60 includes an illumination unit 610 and an imaging unit 620. The illumination unit 610 is disposed on the opposite side of the imaging unit 620 with respect to the target trajectory Q of the droplet target 27. The direction in which the illumination unit 610 and the imaging unit 620 are aligned is perpendicular to the target trajectory Q in FIG. 2 , but may be non-perpendicular. The illumination unit 610 and the imaging unit 620 are attached to a wall 211 outside the chamber 2, the illumination unit 610 is disposed coaxially with a window 218 provided in the wall 211, and the imaging unit 620 is disposed coaxially with a window 219 provided in the wall 211.

[0040] The illumination unit 610 includes a container 611, and a light source 613 and an illumination optical system 615 housed in the container 611. The light source 613 is, for example, a flash lamp that emits light containing multiple wavelengths. The illumination optical system 615 includes a collimator lens. The timing of emission of illumination light emitted from the light source 613 toward the droplet target 27 and the mist target 274 in the plasma generation region R1 is controlled by the CPU 8C.

[0041] The imaging unit 620 includes a container 621, an imaging optical system 623 housed in the container 621, a shutter 625, and an imaging main body 627. The imaging optical system 623 includes a first lens and a second lens. The imaging main body 627 is, for example, a CCD (Charge-Coupled Device).

[0042] When the CPU 8C receives a passing timing signal from the target detection sensor 41, it outputs an imaging trigger signal to each of the shutter 625 and the imaging main unit 627, with a predetermined delay time from the input of the passing timing signal. Hereinafter, the imaging trigger signal for the shutter 625 may be referred to as a shutter trigger signal, and the imaging trigger signal for the imaging main unit 627 may be referred to as an imaging trigger signal. When the shutter 625 receives the shutter trigger signal, it opens for an extremely short time and then closes. The imaging main unit 627 receives the imaging trigger signal and receives illumination light while the shutter 625 is open. The imaging main unit 627 then captures images of the droplet target 27 and the mist target 274 to generate image data and outputs the image data as an electrical signal to the CPU 8C.

[0043] FIG. 3 is a diagram showing the imaging area of ​​the target image capturing section 60. As shown in FIG.

[0044] The droplet target 27 moves from top to bottom in Fig. 3 along the target trajectory Q. The region where the droplet target 27 is turned into mist is called the target mist region Rmist. The positional relationship between the target mist region Rmist, the plasma generation region R1, and the subsequent droplet target 27 is as shown in Fig. 3, for example.

[0045] Although Figure 3 shows only one subsequent droplet target 27 at the time when the mist target 274 is generated, the target image capturing unit 60 may have an imaging area capable of capturing images of multiple droplet targets 27 traveling along the target trajectory Q.

[0046] The plasma generation region R1 is located below (in the -Y direction) the target mist-forming region Rmist and includes a region that overlaps with the target mist-forming region Rmist. The center of the target mist-forming region Rmist is included within the plasma generation region R1.

[0047] FIG. 4 is a diagram illustrating an example of the arrangement of the EUV light sensor 43. FIG. 5 is a diagram illustrating the arrangement of the EUV light sensor 43 shown in FIG. 4 as viewed from the X-axis direction. As shown in FIGS. 4 and 5, the multiple EUV light sensors 43 are composed of, for example, EUV light sensors 43a to 43c. The multiple EUV light sensors 43a to 43c are provided on the wall 211 of the chamber 2 so as to face the plasma generation region R1 from different directions. Each of the multiple EUV light sensors 43a to 43c is arranged so as not to block the optical path of the EUV light 277 reflected by the EUV light collector mirror 231. Each of the multiple EUV light sensors 43a to 43c is arranged along the outer periphery of the EUV light collector mirror 231. The multiple EUV light sensors 43a to 43c are arranged equidistant from each other with respect to the plasma generation region R1 so as to reduce the difference in energy measured by each of the multiple EUV light sensors when the plasma 275 is generated in the plasma generation region R1.

[0048] Each of the multiple EUV light sensors 43a to 43c is disposed at a position that makes it easy to evaluate the center of gravity of the EUV light 277. For example, the multiple EUV light sensors 43a to 43c are disposed at each vertex of an isosceles right triangle as shown in FIG. 4. The right isosceles triangle shown in FIG. 4 is a right isosceles triangle with the midpoint of its long side located in the plasma generation region R1, its apex angle located on the Z axis, and its two short sides aligned along the X axis and Y axis, respectively. The EUV light sensor 43a is an EUV light sensor disposed at a vertex of the right isosceles triangle shown in FIG. 4 that is located on the axis along the Y axis. The EUV light sensor 43b is an EUV light sensor disposed at a vertex of the right isosceles triangle shown in FIG. 4 that is located on the axis along the X axis. The EUV light sensor 43c is an EUV light sensor disposed at a vertex of the right isosceles triangle shown in FIG. 4 that is located on the Z axis.

[0049] The position of the center of gravity of the EUV light 277 is the position of the center of gravity of the energy distribution of the EUV light 277. In other words, the position of the center of gravity of the EUV light 277 is the position of the weighted average in the energy distribution of the EUV light 277. Specifically, the position of the center of gravity of the EUV light 277 is a spatial position identified from a plurality of measurement values ​​obtained by measuring the energy of the EUV light 277 with a plurality of EUV light sensors 43 a to 43 c. The position of the center of gravity of the EUV light 277 is an index reflecting the irradiation position of the main pulse laser beam 31M on the mist-like target 274. The position on the mist-like target 274 where the main pulse laser beam 31M is irradiated is determined according to the position of the center of gravity of the EUV light 277.

[0050] 1.2 Operation FIG. 6 is a flowchart showing an example of the operation of the EUV light generation system 1 according to the comparative example.

[0051] In step S10, the CPU 8C activates each device constituting the EUV light generation system 1 and begins controlling each device so that the desired EUV light 277 is generated. The CPU 8C begins individual control of each device so that the target values ​​preset for each device are met, and also begins an operation of recording data indicating the variation in the distance between adjacent droplet targets 27, data indicating the size of the droplet targets 27, and data indicating the irradiation position of the main pulse laser beam 31M on the mist-like target 274 in the memory 8M in chronological order. Hereinafter, the data recorded in chronological order will be referred to as log data. At this point, the EUV shutter 243 is closed, and the EUV light 277 is not output outside the EUV light generation system 1. The processing of step S10 will be described in detail later ( FIG. 7 ).

[0052] In step S20, the CPU 8C starts monitoring the EUV light generation system 1. The monitored content is the stability of the pulse energy of the EUV light 277. The pulse energy of the EUV light 277 may be referred to as EUV pulse energy. A method for evaluating the stability of the EUV light 277 will be described in detail later.

[0053] In step S30, the CPU 8C opens the EUV shutter 243 and outputs the EUV light 277 to the exposure tool 9.

[0054] In step S40, the CPU 8C makes a determination corresponding to the monitoring started in step S20. That is, the CPU 8C determines whether there is any abnormality in the stability of the EUV light 277. If the determination result shows that there is no abnormality, the processing of step S40 loops. If the determination in step S40 shows that the stability of the pulse energy of the EUV light 277 has fallen below a predetermined value, the CPU 8C determines that an abnormality has occurred in the stability of the EUV light 277, and proceeds to step S50. The details of step S40 will be described later.

[0055] In step S50, the CPU 8C closes the EUV shutter 243 to interrupt the output of the EUV light 277. Note that even when the EUV shutter 243 is closed, the generation of the EUV light 277 continues.

[0056] In step S60, the CPU 8C diagnoses the log data of the EUV light generation system 1 and fine-tunes some devices so that the stability of the pulse energy of the EUV light 277 returns to a predetermined value. After the fine-tuning, the CPU 8C waits until all the log data necessary to determine the stability of the EUV light 277 is available. Details will be described later.

[0057] In step S70, the CPU 8C determines whether or not the abnormality in the stability of the EUV light 277 has been resolved. After performing the fine adjustment in step S60, the CPU 8C generates new monitoring results, and if it is determined in step S70 that the abnormality in the stability of the EUV light 277 has been resolved, the process returns to step S30 and the EUV light 277 is output to the outside.

[0058] On the other hand, if the CPU 8C determines in step S70 that the abnormality in the stability of the EUV light 277 has not been resolved, the process proceeds to step S80.

[0059] In step S80, the CPU 8C shuts down the EUV light generation system 1. After step S80, the flowchart in FIG. 6 ends.

[0060] <Detailed explanation of step S10> Fig. 7 is a flowchart showing details of the processing applied in step S10 of Fig. 6. In the EUV light generation system control start step (step S10) shown in Fig. 7, generation of log data also starts when control starts.

[0061] In step S110, the CPU 8C activates the target supplier 25, the target detection sensor 41, the target image capturing unit 60, and the EUV light sensor 43, and starts the following control.

[0062] The CPU 8C controls the target supplier 25 to output the droplet targets 27 from the target supplier 25 toward the target mist-forming region Rmist. Specifically, the CPU 8C heats the heater 253 of the target supplier 25 to a temperature equal to or higher than the melting point of the target material 257, thereby melting the solid target material 257 contained in the tank 251 of the target supplier 25.

[0063] When the target material 257 is tin, the CPU 8C heats the heater 253 to a temperature of, for example, 250°C to 290°C, since the melting point of tin is 232°C. The CPU 8C controls the pressure regulator 254 of the target supply device 25 to apply a predetermined pressure to the target material 257 in the tank 251 so that the target material 257 in the tank 251 is continuously output from the nozzle 252. Next, the CPU 8C vibrates the piezoelectric element 255 with a predetermined waveform so that the target material 257 output from the nozzle 252 turns into droplets, and multiple droplets combine to generate combined droplets of a predetermined diameter and a predetermined cycle. Hereinafter, the combined droplets will also be referred to as droplet targets 27.

[0064] The droplet target 27 output into the chamber 2 passes through the target detection region R2. The target detection sensor 41 detects the timing when the droplet target 27 passes through the target detection region R2 and generates a passing timing signal. Specifically, the light source 411 of the illumination unit 410 outputs illumination light toward the target detection region R2 via the illumination optical system 412 to illuminate the droplet target 27 passing through the target detection region R2. The optical sensor 421 of the detection unit 420 detects the illumination light output to the target detection region R2 via the light receiving optical system 422, thereby detecting the droplet target 27 passing through the target detection region R2. The light intensity of the illumination light detected by the optical sensor 421 may decrease when the droplet target 27 passes through the target detection region R2. The optical sensor 421 generates a passing timing signal, which is an output signal corresponding to a change in the intensity of the detected illumination light, and transmits the signal to the CPU 8C.

[0065] The CPU 8C determines that the droplet target 27 has passed through the target detection region R2 when the passing timing signal becomes lower than a predetermined threshold. The CPU 8C generates a target detection signal indicating that the droplet target 27 has passed through the target detection region R2 when the passing timing signal becomes lower than the predetermined threshold. The CPU 8C controls the piezo element 255 so that the frequency of the target detection signal becomes a predetermined frequency.

[0066] The droplet target 27 that passes through the target detection region R2 is supplied to the target mist-generation region Rmist. The target image capturing unit 60 generates image data by capturing an image of an area including the target mist-generation region Rmist and transmits the image data to the CPU 8C. The CPU 8C defines a target trajectory Q of the droplet target 27 from the image data and calculates the amount of drive of the stage 26 required for the target trajectory Q to pass through the target mist-generation region Rmist. The CPU 8C then controls the stage 26 based on this drive amount so that the target trajectory Q passes through the target mist-generation region Rmist.

[0067] The CPU 8C also generates log data indicating the variation in the distance between adjacent droplet targets 27 from the image data and records the log data in the memory 8M. The variation in the distance between adjacent droplet targets 27 is recorded as the variation in the time interval (variation in the time difference) σ between the droplet targets 27 reaching the target mist-generation region Rmist. The variation in the time interval σ is defined as follows:

[0068] The variation σ in the time interval between the time when the preceding droplet target 27 reaches the target mist region Rmist and the time when the droplet target 27 traveling immediately thereafter reaches the target mist region Rmist is calculated, for example, using the following equation:

[0069] Time interval variance σ [%] = (σst / μt) × 100 Here, σst is the standard deviation for multiple time intervals included in a unit time. μt is the average value of multiple time intervals included in a unit time. The unit time is several seconds, for example, 1 to 5 seconds. Instead of the time interval variation σ, an index for evaluating the variation in the time intervals of the droplet target 27 may be, for example, n times the standard deviation σst, where n is an arbitrary positive number. The CPU 8C also generates log data indicating the sizes of the droplet target 27 in the X and Y directions from the image data and records the log data in the memory 8M.

[0070] In step S120, the CPU 8C determines whether a predetermined droplet target 27 has been generated. Specifically, the CPU 8C determines whether the droplet target 27 supplied to the target mist-generation region Rmist passes through a predetermined target trajectory Q, and whether the droplet target 27 has reached a predetermined size. If it is determined in step S120 that both conditions are not met, the process of step S120 loops. If it is determined in step S120 that both conditions are met, the process proceeds to step S130.

[0071] In step S130, the CPU 8C starts controlling the pre-pulse laser device 3P. That is, the CPU 8C starts the stages of the pre-pulse laser device 3P and the high-reflection mirror 331, etc., and starts the following control.

[0072] The CPU 8C transmits a trigger signal that triggers the output of the pre-pulse laser light 31P to the pre-pulse laser device 3P at a timing delayed by the delay time Tdp from the timing at which the target detection signal is generated, and causes the pre-pulse laser light 31P to be output.

[0073] The delay time Tdp is a time for matching the timing at which the pre-pulse laser beam 31P is focused in the target mist-forming region Rmist with the timing at which the droplet targets 27 are supplied to the target mist-forming region Rmist. The delay time Tdp is a time calculated by the CPU 8C from the distance between the target mist-forming region Rmist and the target detection region R2, and the frequency of the target detection signal.

[0074] Upon receiving a trigger signal, the pre-pulse laser device 3P outputs a pre-pulse laser beam 31P. The pre-pulse laser beam 31P output from the pre-pulse laser device 3P is reflected by a high-reflection mirror 331P and a combiner 332 of the laser beam transmission optical system 33, passes through a window 215, and is introduced into the chamber 2. The pre-pulse laser beam 31P introduced into the chamber 2 is focused into a target mist-forming region Rmist by the laser beam focusing optical system 22. The pre-pulse laser beam 31P focused into the target mist-forming region Rmist is irradiated onto a droplet target 27 supplied to the target mist-forming region Rmist, generating a mist-like target 274.

[0075] The target image capturing unit 60 generates image data of the mist-like target 274 and transmits the image data to the CPU 8C. The CPU 8C controls the stages of the high-reflection mirror 331P and the combiner 332 and the manipulator 224 of the laser beam focusing optical system 22 to control the irradiation position of the pre-pulse laser beam 31P so that the mist-like target 274 having a predetermined shape is generated from the image data.

[0076] Furthermore, the target image capturing section 60 controls the output of the pre-pulse laser device 3P as follows so that a mist-like target 274 having a predetermined shape is generated.

[0077] The pre-pulse laser beam 31P reflected by the beam splitter 500P is incident on the laser beam sensor 501P, which generates a detection signal. This detection signal is sent to the CPU 8C, which controls the pre-pulse laser device 3P based on the detection signal so that the reflected pre-pulse laser beam 31P has a predetermined energy. Because the transmittance of the beam splitter 500P is known, the CPU 8C controls the energy of the reflected pre-pulse laser beam 31P so that the energy of the transmitted pre-pulse laser beam 31P becomes a predetermined value.

[0078] In step S140, the CPU 8C determines whether or not the mist target 274 of the predetermined shape has been generated. If the CPU 8C determines in step S140 that the mist target 274 of the predetermined shape has not been generated, the process of step S140 loops. If the CPU 8C determines in step S140 that the mist target 274 of the predetermined shape has been generated, the process proceeds to step S150.

[0079] In step S150, the CPU 8C starts controlling the main pulse laser device 3M. That is, the CPU 8C starts up the main pulse laser device 3M, the stage including the high-reflection mirror 331M, and the EUV light sensor 43, and starts the following control.

[0080] The CPU 8C transmits a trigger signal that triggers output of the main pulse laser beam 31M to the main pulse laser device 3M at a timing delayed by the delay time Tdm from the delay time Tdp, and causes the main pulse laser beam 31M to be output.

[0081] The delay time Tdm is a time for matching the timing at which the main pulse laser beam 31M is focused on the plasma generation region R1 with the timing at which the mist-like target 274 moves to the plasma generation region R1. The delay time Tdm defines the timing at which the main pulse laser beam 31M is irradiated onto the mist-like target 274. The delay time Tdm is calculated by the CPU 8C from the distance between the plasma generation region R1 and the target mist-forming region Rmist, and the frequency of the target detection signal.

[0082] The main pulse laser beam 31M is reflected by high-reflection mirrors 331M and 332M of the laser beam transmission optical system 33, passes through the combiner 332 and the window 215, and is introduced into the chamber 2. The main pulse laser beam 31M introduced into the chamber 2 is focused onto the mist-like target 274 in the plasma generation region R1 by the laser beam focusing optical system 22. The mist-like target 274 is converted into plasma, and EUV light 277 is emitted.

[0083] The plurality of EUV light sensors 43a to 43c transmit detection signals of the EUV light 277 to the CPU 8C. The CPU 8C calculates the pulse energy of the EUV light 277 from these detection signals. The CPU 8C controls the stages of the high-reflection mirrors 331M and 332M to control the irradiation position of the main pulse laser beam 31M so that the calculated value becomes a target value for the pulse energy of the EUV light 277. The CPU 8C also controls the output of the main pulse laser device 3M as follows so that the pulse energy of the EUV light 277 becomes the target value.

[0084] The main pulse laser beam 31M reflected by the beam splitter 500M is incident on the laser beam sensor 501M, which generates a detection signal. This detection signal is sent to the CPU 8C, which controls the main pulse laser device 3M based on the detection signal so that the reflected main pulse laser beam 31M has a predetermined energy. Since the transmittance of the beam splitter 500M is known, the CPU 8C controls the energy of the reflected main pulse laser beam 31M so that the energy of the transmitted main pulse laser beam 31M becomes a predetermined value by controlling the energy of the reflected main pulse laser beam 31M. The laser beam sensor 501M is an example of a "pulse laser beam sensor" in this disclosure.

[0085] In order to improve the stability of the pulse energy of the EUV light 277, the CPU 8C controls the irradiation position of the main pulse laser beam 31M on the mist-like target 274 as follows.

[0086] The CPU 8C calculates the position of the center of gravity of the EUV light 277 emitted from the plasma generation region R1. The position of the center of gravity of the EUV light 277 is obtained as follows.

[0087] The CPU 8C transmits a first gate signal to each of the plurality of EUV light sensors 43a-43c at a timing delayed by a predetermined delay time from the timing at which the target detection signal is generated. The first gate signal is a signal that gives each of the plurality of EUV light sensors 43a-43c an opportunity to measure the energy of the EUV light 277. Upon receiving the first gate signal, each of the plurality of EUV light sensors 43a-43c measures the energy of the EUV light 277 and transmits each measurement value to the CPU 8C. The CPU 8C calculates the center of gravity position of the EUV light 277 using a predetermined algorithm.

[0088] The CPU 8C controls the stages of the high-reflection mirrors 331M and 332M based on the calculated center-of-gravity position so as to generate EUV light 277 with stable pulse energy, and controls the irradiation position of the main pulse laser beam 31M on the mist-like target 274. Note that the control amount may be a control amount calculated in advance according to the relative position between the center-of-gravity position and the focal point of the main pulse laser beam 31M.

[0089] The CPU 8C generates log data indicating the relative position of the mist-like target 274 with respect to the focal point of the main pulse laser beam 31M, and records the log data in the memory 8M.

[0090] In step S160, the CPU 8C determines whether or not EUV light 277 with pulse energy of the target value has been generated. If the CPU 8C determines in step S160 that EUV light 277 with pulse energy of the target value has not been generated, the processing of step S160 loops. If the CPU 8C determines in step S160 that EUV light 277 with pulse energy of the target value has been generated, the flowchart of FIG. 7 ends and the processing returns to the flowchart of FIG. 6.

[0091] <Detailed Description of Steps S20 and S40> An EUV pulse energy variation 3σe, which indicates the temporal variation of the EUV pulse energy, is used as an index to monitor the stability of the EUV light 277. The EUV pulse energy variation 3σe is defined, for example, by the following equation:

[0092] EUV pulse energy variation 3σe [%] = (3σse / μe) × 100 Here, σse is the standard deviation of the EUV pulse energy for the plurality of EUV light pulses contained per unit time. μe is the average value of the EUV pulse energy for the plurality of EUV pulses contained per unit time. The pulse energy of the EUV light 277 used to calculate the EUV pulse energy variation 3σe may be, for example, the sum or average of the EUV energy measurements by the EUV light sensors 43a to 43c. The unit time is several seconds, for example, about 1 to 5 seconds. Here, instead of the EUV pulse energy variation 3σe, for example, n times the standard deviation σse may be used as an index for evaluating the stability of the EUV pulse energy.

[0093] If the EUV pulse energy variation 3σe deviates from the allowable range, the CPU 8C determines that the stability of the EUV light 277 is low and that some abnormality has occurred in the EUV light generation system 1.

[0094] <Detailed explanation of step S60> Fig. 8 is a flowchart showing details of the processing applied to step S60 in Fig. 6. In step S60, the CPU 8C diagnoses log data including data before the time when the stability abnormality occurred, and fine-tunes the related configuration based on the diagnosis result so that the pulse energy of the EUV light 277 returns to a stable state.

[0095] In step S610, the CPU 8C diagnoses the log of the binding state of the droplet target 27. The notation "DL" in FIG.

[0096] In step S611, the CPU 8C determines whether or not there is a bonding failure in the droplet target 27. If it is determined in step S611 that there is a bonding failure in the droplet target 27, the process proceeds to step S612.

[0097] In step S612, the CPU 8C adjusts the configuration related to the poor bonding of the droplet target 27. The configuration related to the poor bonding includes the piezoelectric element 255. After the adjustment in step S612, the process proceeds to step S620.

[0098] Furthermore, if it is determined in step S611 that the droplet target 27 is not defectively bonded (if the bonded state is determined to be normal), the process proceeds to step S620.

[0099] In step S620, the CPU 8C diagnoses a log of the variation σ in the time intervals of the droplet targets 27 that are continuously supplied to the target mist-generation region Rmist. The variation σ in the time intervals of the droplet targets 27 is also referred to as the time interval σ of the droplet targets 27, and is represented as "DL time interval σ" in FIG.

[0100] In step S621, the CPU 8C determines whether or not the variation σ in the time intervals of the droplet targets 27 is abnormal. If it is determined in step S621 that the variation σ in the time intervals of the droplet targets 27 is abnormal, the process proceeds to step S622.

[0101] In step S622, the CPU 8C adjusts the configuration related to the variation σ in the time intervals of the droplet target 27. The configuration related to the variation σ in the time intervals includes the piezoelectric element 255. After the adjustment in step S622, the process proceeds to step S630.

[0102] Furthermore, if it is determined in step S621 that there is no abnormality in the variation σ of the time intervals of the droplet targets 27 (if the variation σ of the time intervals is determined to be normal), the process proceeds to step S630.

[0103] In step S630, the CPU 8C diagnoses the relative focusing position of the main pulse laser beam 31M irradiated onto the mist-like target 274. Hereinafter, the relative positional relationship between the focusing position of the main pulse laser beam 31M irradiated onto the mist-like target 274 and the mist-like target 274 will be referred to as the "laser-mist-target relative position."

[0104] In step S631, the CPU 8C determines whether the relative position between the laser and the mist-like target is abnormal. If it is determined in step S631 that the relative position between the laser and the mist-like target is abnormal, the process proceeds to step S632.

[0105] In step S632, the CPU 8C adjusts the configuration related to the relative position of the laser and the mist-like target. The configuration related to the relative position of the laser and the mist-like target includes the stages of the high-reflection mirror 331M and the high-reflection mirror 332M. After the adjustment in step S632, the flowchart in FIG. 8 ends and the process returns to the flowchart in FIG. 6.

[0106] Furthermore, if it is determined in step S631 that there is no abnormality in the relative position between the laser and the mist-like target (if the relative position between the laser and the mist-like target is determined to be normal), the flowchart in FIG. 8 ends and the process returns to the flowchart in FIG. 6.

[0107] 2. Challenges Even if each component of the EUV light generation system 1 is operating normally, an abnormality may occur in the stability of the EUV light 277. It is empirically known that the main causes of an abnormality in the stability of the EUV light 277 are an abnormality in the binding state of the droplet target 27, an abnormality in the time interval σ of the droplet target 27, or an abnormality in the relative position between the laser and the mist-like target.

[0108] However, when it is determined in step S40 of Figure 6 that there is an abnormality in the stability of the EUV light 277, it is difficult for the user to immediately determine which of the factors is the cause. Therefore, the user runs a special diagnostic program optimized for each factor and adjusts the specific configuration based on the diagnostic results. This method requires running all of the diagnostic programs corresponding to each factor, and therefore requires time to return to a state where the EUV light 277 is stably generated.

[0109] 3. Embodiment 1 3.1 Configuration 9 is a diagram showing an EUV light generation system 1A according to embodiment 1. Differences between the EUV light generation system 1A and the comparative example shown in FIG.

[0110] The memory 8M of the processor 8 in the EUV light generation system 1A stores a learning model used in processing to estimate the cause of the stability abnormality of the EUV light 277. This learning model is a trained learning model composed of a neural network created by performing machine learning using training data, so as to input log data on the pulse energy of the main pulse laser beam 31M output by the main pulse laser device 3M, the EUV light pulse energy detected by each of the EUV light sensors 43a to 43c, and the irradiation pulse interval of the main pulse laser beam 31M, and to output the cause of the stability abnormality of the EUV light 277. The learning model is essentially a program.

[0111] The causes of the abnormality in the stability of the EUV light 277 include at least an abnormality in the binding state of the droplet target 27, an abnormality in the time interval σ of the droplet target 27, and an abnormality in the relative position between the laser and the mist-like target.

[0112] A terminal 70 is connected to the processor 8. The terminal 70 has a function of storing a learning model in a memory 8M and a function of displaying factors estimated by the learning model.

[0113] 3.2 Operation Fig. 10 is a flowchart showing an example of the operation of the EUV light generation system 1A according to embodiment 1. Differences between the flowchart in Fig. 10 and Fig. 6 will be described.

[0114] 10, step S10A is included instead of step S10, step S25 is included between step S20 and step S30, step S45 is included between step S40 and step S50, and step S55 is included instead of step S60.

[0115] In step S10A, in addition to the operation of step S10, generation of log data on the pulse energy of the main pulse laser beam 31M output by the main pulse laser device 3M, the EUV light pulse energies detected by the EUV light sensors 43a to 43c, and the irradiation pulse interval of the main pulse laser beam 31M is also started. Each piece of log data is recorded in the memory 8M and updated.

[0116] These three log data can be graphed as shown in Figures 11 to 13. Figure 11 is a graph showing an example of log data of the pulse energy of the main pulse laser beam 31M.

[0117] Fig. 12 is a graph showing an example of log data of the irradiation pulse interval of the main pulse laser beam 31M. The irradiation pulse interval of the main pulse laser beam 31M is the time interval between the passing timing of one preceding target through the target detection region R2 and the passing timing of one subsequent target through the target detection region R2. Fig. 13 is a graph showing an example of log data of EUV light pulse energy detected by one of the EUV light sensors 43a to 43c (for example, the EUV light sensor 43a).

[0118] The horizontal axis of each log data shown in FIGS. 11 to 13 represents time. The time interval for recording data may differ for each log data, but the time range for recording each log data is recorded so that it is the same for each log data. The time range is recorded so that it goes back 3 to 8 seconds from the present. The number of data items that each log data goes back may be 500 to 1500 from the most recent. If the number of data items recorded differs for each log data, the number of data items is adjusted, for example, by zero padding, so that the number of data items for each log data is the same.

[0119] In step S25, the CPU 8C inputs the three pieces of log data whose generation started in step S10A into a trained learning model configured by a neural network, and starts estimating the cause of the stability abnormality of the EUV light 277. Details of step S25 will be described later.

[0120] If it is determined in step S40 that an abnormality has occurred in the stability of the EUV light 277, the process proceeds to step S45. In step S45, the CPU 8C displays the estimated results of the abnormality cause on the terminal 70. Relative values ​​indicating the probability of occurrence of an abnormality in the bonding state of the droplet targets 27, an abnormality in the time interval σ of the droplet targets 27, and an abnormality in the relative position of the laser to the mist-like target are displayed on the terminal 70. A detailed example of the display of the estimated results on the terminal 70 will be described later.

[0121] In step S55, the CPU 8C fine-tunes the configuration associated with the abnormality factor that is most likely to occur so that the stability of the pulse energy of the EUV light 277 returns to a predetermined value. Step S55 will be described in detail later. After the fine-tuning, the CPU 8C continues to detect the EUV light 277 until log data that allows stability to be determined is accumulated.

[0122] After step S55, the process proceeds to step S70. The other steps may be the same as those in FIG.

[0123] <Detailed explanation of step S25> FIG. 14 is a flowchart showing the details of the processing applied in step S25.

[0124] In step S260, the CPU 8C calls the learning model stored in the memory 8M.

[0125] In step S270, the CPU 8C inputs the following three pieces of log data (1) to (3) stored in the memory 8M into the learning model.

[0126] (1) Pulse energy of the main pulse laser beam 31M output from the main pulse laser device 3M (2) Irradiation pulse interval of 31M main pulse laser light (3) EUV light pulse energy detected by each of the EUV light sensors 43a to 43c

[0127] In step S280, the CPU 8C executes calculations using the learning model to estimate the cause of the abnormality. The estimation is executed, for example, every second, and the CPU 8C records the estimation results for each estimation time in the memory 8M.

[0128] After step S280, the flowchart of FIG. 14 ends and the process returns to the flowchart of FIG.

[0129] <Detailed explanation of step S45> In step S45, the CPU 8C displays the estimation result of the abnormality cause on the terminal 70 and identifies the abnormal state with the highest probability. When the CPU 8C determines in step S40 that the EUV light 277 has a stability abnormality, it calls up from the memory 8M the estimation result for the time closest to the determination time from the estimation results stored in the memory 8M and displays it on the terminal 70.

[0130] FIG. 15 is a diagram showing a first example of the display of the estimation results on the terminal 70. FIG. 16 is a diagram showing a second example of the display on the terminal 70. FIG. 15 shows the estimation results of all states at the time when the estimation button 702 is pressed, expressed as probabilities. As shown in FIG. 15, the estimation results using the learning model may be displayed as output values ​​indicating the probability of each of four states: "normal," "abnormal droplet bonding state," "abnormal droplet time interval σ," and "abnormal laser-mist target device position." The output value indicating the probability of each state has a maximum value of 1 and a minimum value of 0. The output value of each state is a relative value, and the sum of the output values ​​of all states is 1. The time when the estimation button 702 is pressed is also displayed as the estimated time.

[0131] When the display switch button 704 on the screen is pressed, the display screen switches to that shown in Fig. 16, which displays the transition of the estimation results for the most recent few seconds for the state showing the highest probability among the four states. Note that when the display switch button 704 on the screen of Fig. 16 is pressed, the display screen switches to that shown in Fig. 15.

[0132] <Detailed explanation of step S55> 17 is a flowchart showing details of the processing applied in step S55. In step S55, the CPU 8C adjusts the specific configuration of the EUV light generation system 1A based on the estimation result in step S45.

[0133] In step S560, the CPU 8C determines whether or not the droplet target 27 is poorly bonded. If the result of the determination in step S560 is that the droplet target 27 is poorly bonded, the process proceeds to step S561. Step S561 is the same as step S612 in FIG. 8.

[0134] If the determination result in step S560 indicates that the droplet target 27 is not poorly bonded, the process proceeds to step S570. In step S570, the CPU 8C determines whether the droplet time interval σ is abnormal. If the determination result in step S570 indicates that the droplet time interval σ is abnormal, the process proceeds to step S571. Step S571 is the same as step S622 in FIG. 8.

[0135] If the droplet time interval σ is determined not to be abnormal in step S570, the process proceeds to step S580. In step S580, the CPU 8C determines whether the laser-mist-like target relative position is outside the stable region. If the laser-mist-like target relative position is determined to be outside the stable region in step S580, that is, if the laser-mist-like target relative position is determined to be abnormal, the process proceeds to step S581. Step S581 is the same as step S632 in FIG. 8. After step S561, step S571, or step S581, the flowchart in FIG. 17 ends and the process returns to the flowchart in FIG. 10.

[0136] Furthermore, if it is determined in step S580 that the relative position between the laser and the mist-like target is within the stable region, the flowchart in FIG. 17 ends and the process returns to the flowchart in FIG.

[0137] For ease of explanation, FIG. 17 shows the steps of step S560, step S570, and step S580 in stages. However, since the most probable abnormality factor is estimated by estimation using the learning model (step S45), the judgments in step S560, step S570, and step S580 can be made all at once by applying the most probable abnormality factor indicated by the estimation result using the learning model.

[0138] 10, the most probable cause of the abnormality is estimated, and therefore the process performed in step S55 is one of droplet combination adjustment (step S561), droplet spacing adjustment (step S571), and laser-mist target device position adjustment (step S581), which corresponds to the most probable cause of the abnormality. For this reason, a diagnostic program is not executed for each cause, as in the comparative example (FIG. 8).

[0139] 3.3 How to generate a learning model 3.3.1 Learning model configuration 18 is a diagram showing the configuration of a learning model used in embodiment 1. The learning model is configured using a deep neural network.

[0140] The learning model includes an input layer, an intermediate layer, and an output layer. Following the first input layer, convolutional layers containing a convolutional network, an activation function, and pooling are placed in the second and third intermediate layers. To improve efficiency during learning, the learning model uses a ReLU function as the activation function in the intermediate layer. Max pooling is applied for pooling. A fully connected layer and an activation function are placed in the output layer. A sigmoid function is used as the activation function for the output layer.

[0141] The second and third layers of the convolutional network detect features, and the output fully connected layer aggregates the features and outputs the final inference results.

[0142] 3.3.2 Learning model generation flow Fig. 19 is a flowchart showing a method for generating a learning model. The flowchart in Fig. 19 will be described below with respect to differences from the flowchart in Fig. 6. It is assumed that the learning model shown in Fig. 18 is stored in the memory 8M.

[0143] The flowchart in FIG. 19 includes steps S64, S65, and S76 between steps S60 and S80 in FIG. 6, and includes step S72 on the way back from step S70 to step S30.

[0144] In step S64 after step S60, the CPU 8C records the combinations of the log data (1) to (3) for the following three abnormal states [1] to [3] diagnosed in step S60 in the memory 8M as training data for the abnormal states. The training data for the abnormal states is an example of the "second training data" in the present disclosure.

[0145] [1] Abnormal binding state of droplet target 27 [2] Anomaly of the time interval σ of droplet target 27 [3] Laser mist target relative position abnormality

[0146] In other words, a combination of log data (1) to (3) that is identified as one of the three abnormal states is recorded as training data for the abnormal state. The training data can be understood as training data for training the learning model.

[0147] In step S65, the CPU 8C determines whether the training data necessary for learning has been prepared sufficiently. Machine learning for generating a learning model requires the preparation of a training data set containing a large amount of training data for abnormal conditions and a large amount of training data for normal conditions. For example, the CPU 8C is configured with the number of training data for abnormal conditions and the number of training data for normal conditions required for learning to achieve desired inference performance, and the CPU 8C determines whether the number of training data collected satisfies the configured conditions.

[0148] If the preparation of the training data is insufficient and the determination result in step S65 is No, the process proceeds to step S70.

[0149] If the determination result in step S70 is Yes, the CPU 8C proceeds to step S72.

[0150] In step S72, the CPU 8C stores the log data (1) to (3) in the normal state in the memory 8M as training data in the normal state. The training data in the normal state is an example of the "first training data" in the present disclosure.

[0151] After step S72, the CPU 8C returns to step S30.

[0152] If the determination result in step S65 is Yes, that is, if all of the teacher data for the abnormal state and the teacher data for the normal state are available, the process proceeds to step S76.

[0153] In step S76, the CPU 8C calls the learning model of FIG. 18 from the memory 8M, inputs all the teaching data into the learning model, causes it to learn, and stores the learned model in the memory 8M.

[0154] The log data to be input to the learning model includes the following three types (1) to (3).

[0155] (1) Pulse energy of the main pulse laser beam 31M output from the main pulse laser device 3M (2) Irradiation pulse interval of 31M main pulse laser light (3) EUV light pulse energy detected by each of the EUV light sensors 43a to 43c

[0156] The learning model is constructed as a four-class classification model that uses the above three types of log data (1) to (3) as input data, estimates which of four states the current state is, and outputs a score (output value) indicating the probability of each of the four states. In other words, the trained model is a neural network trained using training data that associates each of the four states with a set of the three types of log data (1) to (3). The output value indicating the probability of each state is an example of output data output by the learning model.

[0157] After generating the trained model, the process may proceed to step S10A in FIG.

[0158] 3.4 Variations The generation of the learning model described in step S76 is not limited to a configuration in which it is performed by the CPU 8C. For example, the process of step S76 may be performed using an information processing device including a CPU other than the CPU 8C. In this case, the CPU of the other information processing device can receive the training data and generate the learned model.

[0159] Furthermore, the collection of teacher data and the generation of a learning model (machine learning processing) do not need to be performed consecutively, and the collection of teacher data and the generation of a learning model may be performed at separate times.

[0160] 3.5 Actions and Effects According to the EUV light generation system 1A of the embodiment, when an abnormality occurs in the stability of the EUV light 277, the cause of the abnormality is identified using a learning model, so the time required to restore the EUV light generation system 1A from an abnormal state is shorter than in the comparative example.

[0161] 4. Exposure equipment and inspection equipment FIG. 20 is a diagram showing a schematic configuration of an exposure apparatus 9 connected to the EUV light generation system 1A.

[0162] In FIG. 20 , the exposure apparatus 9 as an external device includes a mask irradiation unit 98 and a workpiece irradiation unit 99. The mask irradiation unit 98 illuminates a mask pattern on a mask table MT via a reflection optical system with EUV light 277 incident from the EUV light generation system 1A. The workpiece irradiation unit 99 forms an image of the EUV light 277 reflected by the mask table MT onto a workpiece (not shown) placed on a workpiece table WT via a reflection optical system. The workpiece is a photosensitive substrate such as a semiconductor wafer coated with photoresist. The exposure apparatus 9 exposes the workpiece to the EUV light 277 reflecting the mask pattern by synchronously translating the mask table MT and the workpiece table WT. Electronic devices can be manufactured by transferring a device pattern onto a semiconductor wafer using the exposure process described above.

[0163] An inspection apparatus 91 may be connected to the EUV light generation system 1A as an external apparatus instead of the exposure apparatus 9. Fig. 21 is a diagram showing a schematic configuration of the inspection apparatus 91 connected to the EUV light generation system 1A.

[0164] In FIG. 21 , an inspection apparatus 91 serving as an external device includes an illumination optical system 93 and a detection optical system 96. The illumination optical system 93 reflects EUV light 277 incident from the EUV light generation system 1 and irradiates a mask 95 placed on a mask stage 94. The mask 95 here includes a mask blank before a pattern is formed. The detection optical system 96 reflects the EUV light 277 from the illuminated mask 95 and forms an image on the light-receiving surface of a detector 97. The detector 97 receives the EUV light 277 and acquires an image of the mask 95. The detector 97 is, for example, a TDI (Time Delay Integration) camera. The image of the mask 95 acquired through the above process is used to inspect the mask 95 for defects, and the inspection results are used to select a mask suitable for manufacturing electronic devices. The pattern formed on the selected mask is then exposed and transferred onto a photosensitive substrate using an exposure apparatus 9, thereby manufacturing electronic devices.

[0165] 5.Other The above description is intended to be illustrative rather than limiting. Thus, it will be apparent to one skilled in the art that modifications can be made to the disclosed embodiments without departing from the scope of the claims. It will also be apparent to one skilled in the art that the disclosed embodiments can be used in combination.

[0166] Terms used throughout this specification and claims should be construed as "open ended" unless expressly stated otherwise. For example, words such as "comprise," "have," "comprise," and "equip" should be construed as meaning "without excluding the presence of elements other than those listed." In addition, the modifier "a" should be construed as meaning "at least one" or "one or more." In addition, the term "at least one of A, B, and C" should be construed as "A," "B," "C," "A+B," "A+C," "B+C," or "A+B+C." Furthermore, it should be construed to include combinations of these with elements other than "A," "B," and "C."

Claims

1. 1. An extreme ultraviolet light generation system that generates extreme ultraviolet light by irradiating a droplet target formed by combining a plurality of droplets with a pre-pulse laser beam to generate a mist-like target, and irradiating the mist-like target with a main pulse laser beam to generate plasma, a pulse laser beam sensor that measures the pulse energy of the main pulse laser beam; a target detection sensor that generates a droplet target passing signal, which is used to generate a trigger signal for irradiating the main pulse laser beam; an EUV light sensor that measures pulse energy of the extreme ultraviolet light; a processor, The processor: log data of pulse energy obtained from the pulse laser light sensor, log data of the irradiation pulse interval of the main pulse laser light, and log data of pulse energy obtained from the EUV light sensor are used as input data; a neural network that outputs information that can identify whether the extreme ultraviolet light generation system is in a normal state, a state in which the droplet targets are poorly bonded, a state in which the dispersion of the intervals between the droplet targets is abnormal, or a state in which the relative position between the position irradiated with the main pulse laser beam and the mist-like target is abnormal, Extreme ultraviolet light generation system.

2. 2. The extreme ultraviolet light generating system according to claim 1, Further, a terminal for displaying the information is provided. Extreme ultraviolet light generation system.

3. 3. The extreme ultraviolet light generating system according to claim 2, The terminal displays an output value indicating a probability of the state for each of the states. Extreme ultraviolet light generation system.

4. 4. The extreme ultraviolet light generating system according to claim 3, The output value is a relative value in a range of values, with 1 being the maximum value and 0 being the minimum value, the sum of the output values ​​for each state is 1; Extreme ultraviolet light generation system.

5. 3. The extreme ultraviolet light generating system according to claim 2, the processor executes a process of estimating which of the states occurs using the neural network at specific time intervals; The terminal Display the state with the maximum output value at each estimated time. Extreme ultraviolet light generation system.

6. 2. The extreme ultraviolet light generating system according to claim 1, Each of the log data is data measured during the same period. Extreme ultraviolet light generation system.

7. 7. The extreme ultraviolet light generating system according to claim 6, The period is 3 to 8 seconds. Extreme ultraviolet light generation system.

8. 8. The extreme ultraviolet light generating system according to claim 7, The number of data items in each of the log data periods is 500 to 1500. Extreme ultraviolet light generation system.

9. 9. The extreme ultraviolet light generating system according to claim 8, If the number of data items differs among the log data items, the number of data items is adjusted so that the number of data items in each of the log data items is the same. Extreme ultraviolet light generation system.

10. 9. The extreme ultraviolet light generating system according to claim 8, The number of data items in the respective log data periods is the same. Extreme ultraviolet light generation system.

11. 2. The extreme ultraviolet light generating system according to claim 1, The intermediate layer of the neural network includes a convolutional network. Extreme ultraviolet light generation system.

12. 2. The extreme ultraviolet light generating system according to claim 1, The activation function of the intermediate layer of the neural network is a ReLU function. Extreme ultraviolet light generation system.

13. 2. The extreme ultraviolet light generating system according to claim 1, The intermediate layer of the neural network includes a max pooling process. Extreme ultraviolet light generation system.

14. 2. The extreme ultraviolet light generating system according to claim 1, The output layer of the neural network includes a fully connected layer. Extreme ultraviolet light generation system.

15. 15. The extreme ultraviolet light generating system according to claim 14, The activation function of the output layer of the neural network is a sigmoid function. Extreme ultraviolet light generation system.

16. 2. The extreme ultraviolet light generating system according to claim 1, The neural network is a neural network trained using training data that associates each of the states with each of the log data. Extreme ultraviolet light generation system.

17. 17. The extreme ultraviolet light generating system according to claim 16, the training data includes first training data which is the log data when the extreme ultraviolet light generation system is in a normal state, and second training data which is the log data when the extreme ultraviolet light generation system is in an abnormal state; Extreme ultraviolet light generation system.

18. A method for manufacturing an electronic device, comprising:

1. An extreme ultraviolet light generation system that generates extreme ultraviolet light by irradiating a droplet target formed by combining a plurality of droplets with a pre-pulse laser beam to generate a mist-like target, and irradiating the mist-like target with a main pulse laser beam to generate plasma, a pulse laser beam sensor that measures the pulse energy of the main pulse laser beam; a target detection sensor that generates a droplet target passing signal, which is used to generate a trigger signal for irradiating the main pulse laser beam; an EUV light sensor that measures pulse energy of the extreme ultraviolet light; a processor, The processor: log data of pulse energy obtained from the pulse laser light sensor, log data of the irradiation pulse interval of the main pulse laser light, and log data of pulse energy obtained from the EUV light sensor are used as input data; generating the extreme ultraviolet light by the extreme ultraviolet light generation system, the extreme ultraviolet light generation system including a neural network that outputs information that can identify whether the extreme ultraviolet light generation system is in a normal state, a state in which the droplet targets are poorly bonded, a state in which the dispersion in the spacing between the droplet targets is abnormal, or a state in which the relative position of the position irradiated with the main pulse laser beam and the mist-like target is abnormal; outputting the extreme ultraviolet light to an exposure device; exposing a photosensitive substrate to the extreme ultraviolet light in the exposure apparatus to manufacture an electronic device; A method for manufacturing electronic devices.

19. A method for manufacturing an electronic device, comprising:

1. An extreme ultraviolet light generation system that generates extreme ultraviolet light by irradiating a droplet target formed by combining a plurality of droplets with a pre-pulse laser beam to generate a mist-like target, and irradiating the mist-like target with a main pulse laser beam to generate plasma, a pulse laser beam sensor that measures the pulse energy of the main pulse laser beam; a target detection sensor that generates a droplet target passing signal, which is used to generate a trigger signal for irradiating the main pulse laser beam; an EUV light sensor that measures pulse energy of the extreme ultraviolet light; a processor, The processor: log data of pulse energy obtained from the pulse laser light sensor, log data of the irradiation pulse interval of the main pulse laser light, and log data of pulse energy obtained from the EUV light sensor are used as input data; a neural network that outputs information that can identify whether the extreme ultraviolet light generation system is in a normal state, a state in which the droplet targets are poorly bonded, a state in which the dispersion of the intervals between the droplet targets is abnormal, or a state in which the relative position of the position irradiated with the main pulse laser beam and the mist-like target is abnormal, and irradiating the mask with the extreme ultraviolet light generated by the extreme ultraviolet light generation system to inspect the mask for defects; selecting the mask using the results of the inspection; transferring a pattern formed on the selected mask onto a photosensitive substrate by exposure; A method for manufacturing electronic devices.

Images